System monitoring applications are used to monitor the operational status of the resources made available in a target system. Monitored system resources include CPU, disk, and memory utilization, and application or process status, for example. By monitoring the status of system resources, an operator may take preemptive corrective action and thereby possibly avoid a serious system failure.
Remote monitoring of the status of a system's resources may be implemented out of a desire to consolidate the operations of multiple systems into a single location. In remote monitoring, a system's status information is transmitted via a network (local area or wide area) to a workstation, for example, and displayed thereon. In some monitoring arrangements, software on a single workstation can display the status information of multiple target systems and also provide an interface for operating the target systems. The Operations Sentinel software from Unisys is an example package that provides such monitoring and operations capabilities for multiple target systems. By consolidating the monitoring and operations of multiple systems on a single workstation it is possible for a single user to operate multiple systems from a single location.
During the course of operating a computing arrangement it may be desirable to change the conditions under which status information is reported for various resources, enable a new software component for monitoring additional resources, or remove a current software component when monitoring of a resource is no not longer desired. As new applications are added to the system and new users draw on the system's resources, for example, the operational status of the system may become more critical and administrators may desire more advanced warning of approaching resource shortages.
Changing the parameters involved in monitoring a system, however, may be burdensome and present undesirable risks. In order to change the monitoring capabilities in some systems it may be necessary to stop a monitoring program, change the operational parameters, and thereafter restart the program. However, restarting a monitoring program on the target system may be time-consuming. Furthermore, during the time it takes to restart the monitoring program, important status information may be lost, and the opportunity to take remedial action may have passed.
A method and system that address these and other related issues are therefore desirable.